Publication Date
1997
Description
A system for making a raster map for predicting the heading date of timothy (Phleum pratense L.) at first cutting for every 1km2 plot was developed to supply information for smooth harvest of forage of good quality in a dairy farming area of Hokkaido. Daily mean air temperature for every 1km2 plot was estimated with data from a network of meteorological observatories and data base of the Japan Meteorological Agency. Day length could be calculated from latitude and calendar day. Using these two environmental factors, heading date of timothy at first cutting for each area was predicted by nonparametric DVR method. Prediction errors due to the estimation process for daily mean air temperatures and the prediction process of heading date were less than ±5 days and less than ±2 days, respectively. Therefore, when a difference between predicted heading date and the observed one exceeded 5 days, the other local factors should be considered.
Citation
Saigusa, T; Nakashima, K; Noshiro, N; and Tsutsumi, M, "Raster Map for Prediction of Heading Date of Timothy By NonParametric DVR Method" (2024). IGC Proceedings (1997-2023). 13.
https://uknowledge.uky.edu/igc/1997/session9/13
Included in
Agricultural Science Commons, Agronomy and Crop Sciences Commons, Plant Biology Commons, Plant Pathology Commons, Soil Science Commons, Weed Science Commons
Raster Map for Prediction of Heading Date of Timothy By NonParametric DVR Method
A system for making a raster map for predicting the heading date of timothy (Phleum pratense L.) at first cutting for every 1km2 plot was developed to supply information for smooth harvest of forage of good quality in a dairy farming area of Hokkaido. Daily mean air temperature for every 1km2 plot was estimated with data from a network of meteorological observatories and data base of the Japan Meteorological Agency. Day length could be calculated from latitude and calendar day. Using these two environmental factors, heading date of timothy at first cutting for each area was predicted by nonparametric DVR method. Prediction errors due to the estimation process for daily mean air temperatures and the prediction process of heading date were less than ±5 days and less than ±2 days, respectively. Therefore, when a difference between predicted heading date and the observed one exceeded 5 days, the other local factors should be considered.
